https://doi.org/10.24928/2023/0209

Development of a Machine Learning-Based Labour Productivity Prediction Tool to Practice Lean Construction

Abhay Saini¹ & Albert Thomas²

¹Master’s Student, Department of Civil Engineering, Indian Institute of Technology Bombay, Powai, Mumbai, Maharashtra 400076, India, [email protected], http://orcid.org/0000-0001-5702-5880
²Assistant Professor, Department of Civil Engineering, Indian Institute of Technology Bombay, Powai, Mumbai, Maharashtra 400076, India, [email protected], http://orcid.org/0000-0002-4924-6592

Abstract

The construction industry is a labour-intensive industry. This is one of the reasons why the industry has significant room to incorporate lean principles and reduce waste. Various lean tools can be implemented in construction projects, such as Kanban, JIT and 5S. However, these tools majorly focus on activities at an aggregate level and do not always incorporate sub-activities carried out within a small activity. The productivity of smaller activities (activities that typically span from minutes to hours) is essential to be assessed and controlled to increase the efficiency of overall activity. This paper aims to develop a labour productivity prediction tool based on machine learning principles and lean ideologies to improve the overall productivity of construction activities, considering the productivity of sub-activities. The developed framework is demonstrated by analyzing the productivity of reinforcement activity in a construction project. In the study, inventory wastes are minimized using the prediction from the developed quantitative labour productivity prediction model. An increase of 13.7% in overall productivity is achieved through the implementation of the developed framework

Keywords

Lean construction process, value stream mapping, machine learning, lean theory

Files

Reference

Saini, A. & Thomas, A. 2023. Development of a Machine Learning-Based Labour Productivity Prediction Tool to Practice Lean Construction, Proceedings of the 31st Annual Conference of the International Group for Lean Construction (IGLC31) , 1326-1336. doi.org/10.24928/2023/0209

Download: BibTeX | RIS Format